Systems and Methods for a Personal Agent

This application claims the benefit of, and priority to, U.S. Provisional Patent Application No. 63/641,512, filed on May 2, 2024, the entire contents of which are hereby incorporated herein by reference in their entirety.

The present application relates to systems and methods for a personal agent, and, more specifically, to a system and method for a personal travel agent, which utilizes artificial intelligence.

Most travel planning sites operate by providing users with tools and information to research, compare, and book various aspects of travel, including flights, accommodation, transportation, activities, and more. However, the abundance of choices on travel planning sites can be overwhelming for some users. Sorting through numerous flights, hotels, and activities can lead to decision fatigue and make it challenging to find the best option. Moreover, there are often hidden fees and fine print, limited personalization to individual needs, reliance on limited accessible reviews, and price discrepancies, which make travel planning a tedious and challenging task.

Accordingly, there is a need for an improved travel planning system that can assist a user with personalized recommendations in order to make informed, economical travel plans.

In accordance with aspects of the present disclosure, a system for generating a travel recommendation includes a processor and a memory coupled to the processor, the memory having instructions stored thereon, which when executed by the processor, cause the system to: receive a first user input indicating a travel request; determine user attributes based on the first user input, the user attributes including at least one of user travel preferences, user demographics, or historical user input data; generate a travel recommendation based on an output of a neural network, the neural network configured to predict relevance scores for a plurality of inventory items based on the user attributes; display the travel recommendation including at least one inventory item selected based on the output of the neural network; and automatically initiate a travel booking based on the travel recommendation.

In an aspect of the present disclosure, the user attributes may be determined by applying natural language processing to the first user input to extract at least one of intent indicators, contextual information, or preference-related keywords. The user attributes may be stored in a profile database for generating future travel recommendations.

In another aspect of the present disclosure, the first user input may be received via a conversational user interface configured to accept at least one of voice input, text input, or image input.

In yet another aspect of the present disclosure, the instructions, when executed by the processor, may further cause the system to: receive a second user input indicating a modification request, the modification request related to the selected at least one inventory item; and modify the travel recommendation based on the second user input.

In a further aspect of the present disclosure, generating the travel recommendation may include: generating a plurality of vectors, each vector corresponding to a feature of the at least one inventory item; applying a weighting factor to each vector based on the user attributes to produce a weighted score, the weighted score indicating a relevance of the feature to a corresponding user attribute; and combining weighted scores to generate a relevance score for the at least one inventory item. The at least one inventory item may be selected for inclusion in the travel recommendation if the relevance score exceeds a predefined threshold.

In yet a further aspect of the present disclosure, the at least one inventory item may include at least one of a flight, a hotel, a travel activity, a dining reservation, or a transportation service.

In an aspect of the present disclosure, the instructions, when executed by the processor, may further cause the system to output an alert indicating confirmation of the travel booking, the alert including at least one of an e-mail, short message service (SMS) message, an in-application notification, or push notification.

In another aspect of the present disclosure, the system may further include a plurality of agents, each agent configured to generate a portion of a travel recommendation within a distinct category of a plurality for travel categories including at least one of lodging, transportation, dining, entertainment, or activities.

In yet another aspect of the present disclosure, each agent may include a skin associated with custom traits including at least one of a vocabulary, a layout, or an interface tools. The skin may be based on the distinct category of the agent.

In a further aspect of the present disclosure, each inventory item may be represented as a vector. Each vector may be updated in real time based on at least one of current availability or pricing of an associated inventory item.

In accordance with aspects of the present disclosure, a method for generating a travel recommendation includes: receiving a first user input indicating a travel request; determining user attributes based on the first user input, the user attributes including at least one of user travel preferences, user demographics, or historical user input data; generating a travel recommendation based on an output of a neural network, the neural network configured to predict relevance scores for a plurality of inventory items based on the user attributes; displaying the travel recommendation including at least one inventory item selected based on the output of the neural network; and automatically initiating a travel booking based on the travel recommendation.

In an aspect of the present disclosure, determining the user attributes may include applying natural language processing to the first user input to extract at least one of intent indicators, contextual information, or preference-related keywords. The user attributes may be stored in a profile database for generating future travel recommendations.

In another aspect of the present disclosure, the method may further include: receiving a second user input indicating a modification request, the modification request related to the selected at least one inventory item; and modifying the travel recommendation based on the second user input.

In yet another aspect of the present disclosure, the neural network may generate the output by: generating a plurality of vectors, each vector corresponding to a feature of the at least one inventory item; applying a weighting factor to each vector based on the user attributes to produce a weighted score, the weighted score indicating a relevance of the feature to a corresponding user attribute; and combining weighted scores to generate a relevance score for the at least one inventory item. The at least one inventory item may be selected for inclusion in the travel recommendation if the relevance score exceeds a predefined threshold.

In a further aspect of the present disclosure, displaying the travel recommendation may include displaying at least one of a flight, a hotel, a travel activity, a dining reservation, or a transportation service.

In an aspect of the present disclosure, the method may further include outputting an alert indicating confirmation of the travel booking, the alert including at least one of an e-mail, short message service (SMS) message, an in-application notification, or push notification.

In another aspect of the present disclosure, generating the travel recommendation may include retrieving a plurality of agents, each agent configured to generate a portion of the travel recommendation within a distinct category of a plurality for travel categories including at least one of lodging, transportation, dining, entertainment, or activities.

In yet another aspect of the present disclosure, retrieving the plurality of agents may include determining a skin for each agent, the skin associated with custom traits including at least one of a vocabulary, a layout, or an interface tools. The skin may be based on the distinct category of the agent.

In a further aspect of the present disclosure, generating the travel recommendation may include representing each inventory item as a vector. Each vector may be updated in real time based on at least one of current availability or pricing of an associated inventory item.

In accordance with aspects of the present disclosure, a non-transitory computer readable storage medium includes instructions that, when executed by a computer, cause the computer to perform a method for digital rights management, the method including: receiving a first user input indicating a travel request; determining user attributes based on the first user input, the user attributes including at least one of user travel preferences, user demographics, or historical user input data; generating a travel recommendation based on an output of a neural network, the neural network configured to predict relevance scores for a plurality of inventory items based on the user attributes; displaying the travel recommendation including at least one inventory item selected based on the output of the neural network; and automatically initiating a travel booking based on the travel recommendation.

The present application relates to systems and methods for a personal agent, and, more specifically, to a system and method for a personal travel agent, which utilizes artificial intelligence.

For the purpose of promoting an understanding of the principles of the present disclosure, reference will now be made to exemplary embodiments illustrated in the drawings, and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the present disclosure is thereby intended. Various alterations, rearrangements, substitutions, and modifications of the features illustrated herein, and any additional applications of the principles of the present disclosure as illustrated herein, which would occur to one skilled in the relevant art and having possession of this disclosure, are to be considered within the scope of the present disclosure.

As used herein, the term “travel agent” includes a computer program, software application and/or software module designed to assist users in booking travel-related products and/or services, such as flights, hotels, rental cars, and vacation packages. For example, a travel agent may utilize various algorithms, databases, and/or interfaces, which may provide users with options, pricing information, and booking capabilities. The travel agent may provide automated assistance to users by analyzing their preferences, budget, and travel requirements to offer suitable options, and/or may retrieve and display relevant travel information such as flight schedules, hotel amenities, and destination guides to help users make informed decisions. For example, the travel agent may incorporate personalization features to tailor recommendations based on users' past booking history, preferences, and search behavior. The travel agent may be accessible through websites, mobile apps, messaging platforms, and/or chatbots, which may be designed to offer a user-friendly, flexible experience. The travel agent and/or travel system platform may prioritize user privacy and data security by implementing robust measures to protect users' personal information and sensitive data collected during the recommendation process.

As used herein, the term “personnel travel administrator” includes a computer program, software application and/or software module designed to manage and/or streamline various aspects of travel arrangements and logistics. For example, the personnel travel administrator may streamline the process of applying for entry into/exit from a country, including various visas, thereby ensuring efficiency and compliance with various state policies. The personnel travel administrator may also automatically monitor and/or dynamically update the status of various documentation such passports to ensure user information is up to date in advance of traveling. If documentation (e.g., visa or passport) needs to be renewed, the personnel travel administrator may streamline the process of gathering data, generating, and/or submitting required documentation.

As used herein, the term “visual promoter” includes a computer program, software application and/or software module designed to promote products, services, and/or brands using visual content. For example, a visual promoter may leverage various technologies to create, distribute, and/or display visual promotional materials across different platforms and channels, such as a travel system. The visual promoter may employ data analytics and machine learning algorithms to analyze user behavior, preferences, and demographics, enabling targeted advertising campaigns tailored to specific audiences. For example, the visual promoter may incorporate personalization features to customize promotional content based on user interactions, purchase history, and preferences, enhancing engagement and conversion rates. The visual promoter may monitor and analyze the performance of visual promotional campaigns using metrics such as impressions, clicks, conversions, and engagement metrics to optimize future campaigns.

As used herein, the term “itinerary builder” includes a computer program, software application and/or software module designed to create detailed plans for their trips, such as gathering, organizing, and presenting information about various travel-related activities and arrangements. For example, the itinerary builder may provide comprehensive information about a chosen destination, including points of interest, attractions, landmarks, restaurants, hotels, transportation options, and/or local events. The itinerary builder may generate and/or display interactive maps to visualize the locations of various attractions, accommodations, and activities, helping users understand the geographical layout of their itinerary. The itinerary builder may provide real-time updates and notifications regarding changes to travel plans, such as flight delays, cancellations, or itinerary adjustments, ensuring users stay informed throughout their trip. Further, the itinerary builder allow users to share their itineraries with travel companions, friends, or family members, facilitating coordination and communication during the trip planning process. The itinerary builder may offer offline access to users, allowing them to access their travel plans and information even without an internet connection, useful for when they are traveling to remote or unfamiliar destinations.

As used herein, the term “travel recommendation engine” includes a computer program, software application and/or software module designed to analyze user preferences, demographics, historical data, and/or contextual information to provide personalized travel recommendations. The travel recommendation engine may leverage machine learning, artificial intelligence, and data analytics techniques to suggest destinations, accommodations, activities, and travel itineraries that align with users' interests and preferences. For example, the travel recommendation engine may generate and/or utilize user profiles based on demographic information, travel history, preferences, interests, and behavior patterns collected from user interactions with a travel system. The travel recommendation engine may analyze a vast amount of travel-related content, including user reviews, ratings, travel guides, blogs, and social media posts, to extract insights and identify relevant recommendations.

Further, the travel recommendation engine may utilize collaborative and/or content-based filtering, which compares user preferences and behaviors to match users' stated preferences, such as destination type, activities, budget, and travel dates. In doing so, the travel recommendation engine may consider contextual factors such as current location, weather conditions, local events, and travel trends to provide timely and relevant recommendations tailored to users' immediate needs and interests. The travel recommendation engine may adapt and refines recommendations based on user feedback, interactions, and/or changes in preferences over time, ensuring that recommendations remain relevant and up-to-date. For example, the travel recommendation engine may monitor and evaluate the performance of recommendations using metrics such as click-through rates, conversion rates, and user engagement, optimizing its algorithms to improve recommendation accuracy and effectiveness.

As used herein, the term “dynamic packager” includes a computer program, software application and/or software module designed to create custom travel packages by dynamically combining various travel components such as flights, accommodations, activities, and transfers based on user preferences, availability, and pricing data. The dynamic packager may leverage algorithms, APIs, and data integrations to assemble personalized travel packages in real-time, while ensuring compliance with supplier and/or vendor policies for providing discounts. The dynamic packager may utilize pricing algorithms and predictive analytics to optimize the pricing of travel packages based on factors such as demand, seasonality, inventory levels, and competitor pricing, offering competitive rates to users. For example, the dynamic packager may offer bundled discounts and/or promotions for booking multiple travel components together as part of a package, providing cost savings and incentives for users to purchase bundled offerings and/or may employ cross-selling and upselling techniques to suggest additional services or upgrades to enhance the travel experience, such as airport transfers, travel insurance, guided tours, or premium accommodations. The dynamic packager may also allow users to mix and match different travel components to create flexible packages that meet their specific needs and preferences, enabling them to build their ideal itinerary.

As used herein, a “user” includes an entity, which can be a human, an organization, a group, and/or automated system, or any other identifiable entity, that interacts with a computer system, software application, a piece of software acting on behalf of another entity, and/or technology platform to perform actions, access resources, and/or receive information. The user typically has a unique identifier (e.g., username or ID) and may have associated permissions or privileges governing their interactions with the system.

As used herein, “travel” products and/or services may include may include a broad spectrum of offerings aimed at meeting the needs and/or preferences of travelers, enhancing their overall travel experiences, and ensuring convenience, comfort, and safety throughout their journeys, such as transportation, accommodations, travel agencies, tours and activities, travel insurance, travel technology, visa and documentation services, travel accessories and gear, dining and food services, and/or currency exchange and banking services. Further, travel products and/or services may include various lifestyle offerings designed to cater to unique preferences, values, and/or aspirations, such as those offerings contributing to the cultivation of their desired way of life and personal expression. The travel products and/or services may be utilized for various purposes, including recreation, tourism, exploration, business, migration, commuting, and/or lifestyle related needs.

The present disclosure may be utilized by and/or incorporated into the Simplenight® platform, including the Global Experience Platform® (GEP) and/or Travel and Lifestyle platform. Simplenight® is a global technology company building innovative enterprise solutions including customizable bookability, cloud-based distribution, dynamic packaging, and merchandising, which delivers ancillary revenue and increased customer loyalty for its partners. It will be understood that the technology in the present disclosure is configured to operate on a variety of devices, as described herein.

The disclosure herein addresses the challenge of creating an adaptive, interactive, and personalized travel booking experience through a conversational interface, which understands natural language inputs, offers tailored recommendations, and simplifies the process of planning and booking travel arrangements. In doing so, the disclosure herein provides a seamless interface for inputting travel preferences, receiving personalized options, and refining plans based on user feedback.

Further, the disclosure herein exemplifies a modular approach, wherein an AI Agent dynamically interacts with specialized components, thereby allowing for flexible adaptation and scaling of services. The various components may include a frontend web client, API gateway services, a user profile, an AI agent, a natural language processing engine and/or large language model, a text-to-speech engine, a booking API, an itinerary generator, and/or a chat session database.

The frontend web client (UI) may serve as the primary interface for users, accepting both text and voice inputs, and displaying system responses, e.g., designed for user-friendliness and accessibility, facilitating the initial user-system interaction. The API gateway (AG) may be the intermediary, routing requests and responses between the UI and backend services, thereby ensuring efficient traffic management, security (e.g., via authentication through the User Profile), and scalability. The user profile (UP) may manage user authentication and stores preferences, enabling personalized interactions, and may enhance security by authenticating user identities and enables the AI Agent to tailor responses and recommendations. The AI Agent (AIA) may be the core intelligence of the system, directing the flow of information and decisions, processing user queries via NLP, managing interactions with the Booking API and Itinerary Generation, and using feedback for service improvement. The Natural Language Processing & Large Language Model (NLP) may interpret user inputs, extracting intent and relevant data, which is crucial for enabling the system to understand and process natural language queries and feedback. The Text-to-Speech Engine (TTS) may convert text responses into audio, providing a more interactive and accessible user experience that bridges the gap between textual data and auditory feedback. The Booking API (BA) may connect the system to external travel services, retrieving inventory options like hotel bookings based on user queries, which are essential for providing real-time, relevant travel options to users. The Itinerary Generation (IG) may compile user preferences and/or bookings into a coherent itinerary, transforming individual selections into a structured travel plan. The Chat Session Database (CSD) may collect user interactions and feedback, directly informing the continuous refinement of the NLP model, providing a feedback loop that is fundamental for the system's ability to learn and improve over time.

Referring to, there is shown an illustration of an exemplary systemfor using a personal agent in accordance with aspects of the present disclosure. The systemincludes one or more client computer systems,, a cloud system, a network, one or more mobile devices, one or more Internet of things (IOT) devices,, a server, and/or system. The client computer systems,communicate with the serveracross the network. In aspects, multiple serversmay be used in a distributed architecture and/or in a cloud.

The networkmay be wired or wireless, and can utilize technologies such as Wi-Fi, Ethernet, Internet Protocol, 4G, and/or 5G, or other communication technologies. The networkmay include, for example, but is not limited to, a cellular network, residential broadband, satellite communications, private network, the Internet, local area network, wide area network, storage area network, campus area network, personal area network, or metropolitan area network.

As will be described in more detail below, the cloud systemmay implement statistical models and/or machine learning models (e.g., neural network) that process the collected data to identify potential threat behaviors. The term “machine learning model” may include, but is not limited to, neural networks, recurrent neural networks (RNN), generative adversarial networks (GAN), decision trees, Bayesian Regression, Naive Bayes, nearest neighbors, least squares, means, and support vector machine, among other data science and machine learning techniques which persons skilled in the art will recognize.

The illustrated networked environment is merely an example. In embodiments, other systems, servers, and/or devices not illustrated inmay be included. In embodiments, one or more of the illustrated components may be omitted. Such and other embodiments are contemplated to be within the scope of the present disclosure.

Referring now to, exemplary components of the controllerare shown. The controllergenerally includes a storage or database, one or more processors, at least one memory, and a network interface. In aspects, the controllermay include a graphical processing unit (GPU), which may be used for processing machine learning network models.

The databasecan be located in storage. The term “storage” may refer to any device or material from which information may be capable of being accessed, reproduced, and/or held in an electromagnetic or optical form for access by a computer processor. Storage may be, for example, volatile memory such as RAM, non-volatile memory, which permanently holds digital data until purposely erased, such as flash memory, magnetic devices such as hard disk drives, and optical media such as a CD, DVD, Blu-ray Disc™M, or the like.

In aspects, data may be stored on the controller, including, for example, user accounts, permissions, licensing documentation, and/or other data. The data can be stored in the databaseand sent via the system bus to the processor. The databasemay store information in a manner that satisfies information security standards and/or government regulations, such as Systems and Organization Controls (e.g., SOC), General Data Protection Regulation (GDPR), and/or International Organization for Standardization (ISO) standards.

As will be described in more detail later herein, the processorexecutes various processes based on instructions that can be stored in the at least one memoryand utilizing the data from the database. With reference also to, a request from a user device, such as a mobile device or a client computer, can be communicated to the controllerthrough the network interface. The illustration ofis exemplary, and persons skilled in the art will understand that other components may exist in controller. Such other components are not illustrated for clarity of illustration.

With reference to, a block diagram for a machine learning networkfor classifying data in accordance with some aspects of the disclosure is shown. In some systems, a machine learning networkmay include, for example, a convolutional neural network (CNN), a regression and/or a recurrent neural network. A deep learning neural network includes multiple hidden layers. As explained in more detail below, the machine learning networkmay leverage one or more classification models(e.g., CNNs, decision trees, a regression, Naive Bayes, k-nearest neighbor) to classify data. In aspects, the classification modelmay use a data fileand labelsfor classification. The machine learning networkmay be executed on the controller(). Persons of ordinary skill in the art will understand the machine learning networkand how to implement it.

In machine learning, a CNN is a class of artificial neural network (ANN). The convolutional aspect of a CNN relates to applying matrix processing operations to localized portions of data, and the results of those operations (which can involve dozens of different parallel and serial calculations) are sets of many features that are delivered to the next layer. A CNN typically includes convolution layers, activation function layers, deconvolution layers (e.g., in segmentation networks), and/or pooling (typically max pooling) layers to reduce dimensionality without losing too many features. Additional information may be included in the operations that generate these features. Providing unique information, which yields features that give the neural networks information, can be used to provide an aggregate way to differentiate between different data input to the neural networks.

Referring to, generally, a machine learning network(e.g., a convolutional deep learning neural network) includes at least one input layer, a plurality of hidden layers, and at least one output layer. The input layer, the plurality of hidden layers, and the output layerall include neurons(e.g., nodes). The neuronsbetween the various layers are interconnected via weights. Each neuronin the machine learning networkcomputes an output value by applying a specific function to the input values coming from the previous layer. The function that is applied to the input values is determined by a vector of weightsand a bias. Learning, in the deep learning neural network, progresses by making iterative adjustments to these biases and weights. The vector of weightsand the bias are called filters (e.g., kernels) and represent particular features of the input (e.g., a particular shape). The machine learning networkmay output logits. Although CNNs are used as an example, other machine learning classifiers are contemplated.